10 BULLETIN 988, U. S. DEPT. OF AGRICULTURE. 
engaged in the excessive activity of heat production at any one time 
are only a small part of the total colony, the rest of them being 
crowded together in the shell of the cluster or in empty cells of the 
honey comb or standing quietly. The amount of work done by the 
bees that are really active is comparable with that done by the man 
in unusual conditions, and is therefore relatively enormous ; and this 
is maintained not only for short periods but through the whole day 
and the whole winter. 
Moreover, it will also be observed that the temperature conditions 
during this experiment were those in which bees are the least active. 
In fact, as mentioned previously, the temperature in the respiration 
chamber during the experiment was about the same as that which 
beekeepers usually maintain in cellars for wintering bees. Colonies 
wintered outdoors, especially if unprotected, must endure in many 
cases much more severe temperature conditions. Furthermore, this 
experiment was conducted at a time of the year when bees are 
naturally more nearly quiescent. Bees are usually more active dur- 
ing the latter part of winter than during late fall and early winter. 
The figures obtained in this experiment, therefore, represent about as 
low an expenditure of energy as is ever found in a colony of bees, 
except for short intervals. In a preliminary test with this colony 
the quantities of carbon dioxid measured were decidedly larger than 
these, owing to less favorable conditions. 
A hygrometer suspended in the chamber was read at frequent 
intervals. The maximum and minimum readings for each day 
are shown in the table. During the first five days the humidity was 
allowed to remain at a high level. This was accomplished by keep- 
ing the air of the system in circulation only part of the time, 
virtually every other hour. During the other five days the humidity 
was kept much lower by maintaining a constant circulation of air 
through the sulphuric acid. There was a very noticeable difference 
in the quantities of water vapor removed from the chamber in the 
several days of the two periods, owing to the fact that the relative 
dryness of the air in the later period was causing a loss of water 
from the wood of the hive. No difference in the activity of the bees 
that could be ascribed to the difference in water- vapor content of the 
air was noticeable in the temperature curves or in the carbon-dioxid 
output of the various days. 
The barometer was read at noon each day. There were no sig- 
nificant changes in barometric pressure during the course of the ex- 
periment. The reading on the 13th was 755 millimeters, which rose 
each succeeding day to 769 on the 16th, then fell to 750 on the 18th. 
It was 767 on the 19th and for the rest of the experiment remained 
within 4 millimeters of this pressure. 
